Problems applying 5.0.x patches

[Steve Brueggeman]

I'm sure it's something I'm doing wrong, but I cannot figure out what!

I'm trying to apply the aic-5.0.x.patch to a clean (newly downloadled from
sunsite) kernel linux-2.0.33.tar.gz.  I have the same problem if I just use the
tarball and overwrite the original kernel sources with tar -xvzf
aic7xxx-5.0.8.tar.gz.  I have tried versions 5.0.5, 5.0.6, and 5.0.8, and all
have the same problem.

I have two Adaptec AHA2940UW controllers, 
Controller 0: has 6 ultra-wide drives (TID's 0-5, & 8) and a Tape (TID 6)
Controller 1: has 2 ultra-wide drives (TID's 9 & 10)

An unadorned linux-2.0.33.tar.gz kernel works fine (except under heavy loads,
where there appears to be an interrupt problem (mostly guessing), which causes
kernel panics in the swapper (usually)).

The patch appears to apply cleanly, and the compile goes well, but when I boot,
(used aic7xxx=verbose on the lilo prompt), I get something like the following
errors scrolling on the screen.  I did a diff against kernel sources with the
patch applied, and a kernel with sources over-written using tar -xzf
aic7xxx-5.0.8.tar.gz, and there were no differences.  There were all the
expected differences when a diff was done on the patched sources, and original
sources.

The only scsi related kernel-module I have is the scsi-tape, and scsi-generic.
All other scsi pieces are compiled into the kernel. 

(scsi1:-1:-1:-1) Bad scp ptr 16 during SELTO
(scsi1:-1:-1:-1) Referenced SCB 255 not valid during SELTO
   SCSISEQ=0x5a SEQADDR=0x8 SSTAT0=0x10 SSTAT1=0x8A

There were other lines (about 6) that kept scrolling too, but they were not on
the screen long enough for me to write them down.  I had to write by hand, and
things were scrolling fast enough to make getting these other lines very
difficult.

These errors happen after about a 5 second pause after controller 1 is reset,
which is my reset delay value.

Anyone know what, if anything I'm doing wrong?



To Unsubscribe: send mail to majordomo at FreeBSD.org
with "unsubscribe aic7xxx" in the body of the message